238 PLAIN CONCRETE 



in setting in air, and that the shrinkage will be two to three 

 times as great with neat cement. The shrinkage in concrete 

 will be much less than with neat cement or cement mortar. 



The shrinkage of concrete is lessened by embedding in it 

 steel rods or bars, as these, by their tensile resistance, prevent 

 the shrinkage of the material in setting. By the experi- 

 ments of Considere, it is found that with 1-3 mortar rein- 

 forced with steel the shrinkage in setting is about one-fifth 

 that of the same mortar without the steel reinforcement. 



Effect of Thermal Changes in Concrete. Nearly all 

 materials expand slightly as they become heated. Con- 

 crete and steel also follow this law. The contraction or the 

 expansion of concrete due to changes in temperature is about 

 the same as that of steel. The average coefficient of expan- 

 sion of a 1-2-4 concrete for each Fahrenheit degree in change 

 of temperature is .0000055. Experiments made on 1-3-6 

 concrete give a coefficient of expansion of .0000065, which 

 is practically the same as the coefficient of steel. 



Effect of Vibration on Concrete. The effect of constant 

 vibration on concrete structures has not been definitely 

 determined. Many buildings and bridges constructed of 

 concrete reinforced with steel rods and bars have withstood 

 heavy and constant vibration, either continuous or inter- 

 mittent, for an extended period of years with no apparent 

 deterioration in strength. Fresh concrete is always, how- 

 ever, subject to deterioration by vibration, and the strength 

 of concrete subjected to jar or shock when setting is materially 

 reduced, because the process of crystallization between the 

 particles, and the consequent cohesion of the mass, seems 

 to be partly destroyed. 



WORKING STRESSES AND STRENGTH VALUES OF 

 CONCRETE 



The ultimate strength of concrete varies so with the pro- 

 portion of the mixture, manner of working, character of 

 ingredients, and age of material, that it is necessary to 

 assume low unit working stresses for it. 



The usual working stress for plain concrete under com- 

 pression is from 250 to 300 Ib. per sq. in., although, in masses, 



